In gambling and economics, the favourite-longshot bias is an observed phenomenon where on average, betters tend to overvalue "longshots" and relatively undervalue favourites.
In the long run, losing 5% by betting on the favourite, but losing 40% on longshots is not uncommon.
The phenomenon was first discovered by Griffith.
[1] Various theories exist to explain why people willingly bet on such losing propositions, such as risk-loving behavior, risk-averse behavior[2] or simply inaccurate estimation as presented by Sobel and Raines.
[3] Methods such as the goto_conversion,[4] Power[5] and Shin[6] can be used to measure the bias by converting betting odds to true probabilities.